Project Summary/Abstract Most human cancers show evidence of genome-wide epigenetic lesions. One common epigenetic alteration is loss of imprinting (LOI) which is defined as either biallelic expression or silencing of normally monoallelically expressed genes. Recently, several studies have provided compelling evidence that LOI has a causal role in cellular transformation. A large proportion of patients with Beckwith-Wiedemann syndrome (BWS), an imprinting disorder and cancer predisposition condition, exhibit loss of methylation (LOM) at the differentially methylated region in human chromosome 11p15.5 known as KvDMR1. This epimutation, which is also observed in several adult cancers, is associated with LOI (i.e., silencing) of the tumor suppressor gene CDKN1C. During the last 19 month grant period, we have completed studies that show that the KvDMR1 imprinting control region (ICR) contains both the promoter for the Kcnq1ot1 noncoding RNA (ncRNA) and sequences possessing repressive activity in enhancer-blocking assays. Moreover, we have shown that the epigenetic regulatory protein CTCF binds to the minimal repressive sequence in an allele-specific manner. We have also completed studies on our Kcnq1ot1 truncation mouse model which has not only demonstrated a critical role for the ncRNA in regulating imprinted expression, but has uncovered a second silencing mechanism specific for the tumor suppressor gene Cdkn1c. In combination with a Cdkn1c null allele, this truncation mouse has been instrumental in generating a novel mouse model that recapitulates several aspects of BWS. The Kcnqot1 truncation mouse has also allowed us to demonstrate that, while expression of the ncRNA is required for the establishment of nonequivalent nuclear localization of the KvDMR1 imprinted domain, it is dispensable later in development. The current proposal consists of 2 specific aims intended to further our understanding of the mechanism(s) of KvDMR1 function: (1) The CTCF binding sites in KvDMR1 will be deleted either alone or in combination with the Kcnq1ot1 truncation. These studies will determine the role for CTCF binding at this locus; (2) Expression of full length Kcnq1ot1 will be conditionally regulated at fertilization, and the Kcnq1ot1 promoter will be deleted at time points corresponding to the different developmental stages where paternal gene silencing takes place; these analyses will determine whether expression of the ncRNA is required during gametogenesis, and whether the ncRNA is necessary for the maintenance of imprinted expression once it is established. Since other imprinted domains containing growth regulating genes may be controlled in a similar fashion as the KvDMR1 subdomain, and it is likely that there are many other CTCF mediated chromatin insulators and ncRNAs in the mammalian genome, information derived from these studies will have widespread relevance to the epigenetics of cancer.